Tobias Achterberg

Just to make sure I understand it correctly: you get these very strange values like 1e+69 or 1e-310 in the model, even though you never added those coefficients in your Python program? My guess is ...

In order to count the number of non-zeros, you would typically introduce binary variables that indicate whether a given variable is 0 or not. So, if you have integer variables x1,...,xn, then you w...

Your expectations are correct. An overall model status of OPTIMAL means that there was at least one feasible scenario, but there can also be infeasible scenarios. Whether a particular scenario is f...

1) Correct, you cannot set variable attributes in a callback. But you can produce the cut "x <= 0" and add this as a user cut. Internally, this will then be converted into a bound change. 2) Yes, i...

Yes, it is not possible to pass an LP basis to a MIP solve. This is mostly because the MIP preprocessing would usually invalidate any warm start basis anyways, because most of the MIP preprocessing...

I would like to investigate. Could you please send me - your original model as *.mps file, - your MIP start as *.mst file or *.sol file, and - your modified model with bounds fixed to the MIP start...

Where do these log files come from? From the feasRelax model?

I don't understand why you want to use a callback function here. Inside a callback, you cannot modify or extend the model that you are currently solving. It sounds more like you want to have an out...

That is not a particularly large model. But it is hard to say how you can improve your model without knowing anything about it. Would it be possible to send me your *.mps file? I would like to take...

Yes, using an indicator is the right way to go. And it is very easy, you just need to invert the direction of the indicator. Instead of f_r > 0 -> i_r = 1 you would enforce i_r = 0 -> f_r <= 0 This...